In [1]:
import numpy as np
import pandas as pd
import ast
from pandas import Series, DataFrame
import matplotlib.pyplot as plt 
from matplotlib.ticker import StrMethodFormatter
from itertools import combinations
from scipy import stats
import seaborn as sns
import statsmodels.api as sm
from sklearn import linear_model as lm
from scipy.misc import factorial
import matplotlib
%matplotlib notebook

Load Global Terrorism Database an filter colums

In [2]:
df = pd.read_csv('globalTerrorism.csv', engine = 'python')
dataset1 = pd.concat([df['country'],df['country_txt'],df['region'],df['provstate'],df['city'],df['targtype1'],df['attacktype1'],df['nkill'],df['nwound']] ,axis=1)
happiness = pd.read_csv('World-Happiness-Report/2017.csv', engine = 'python')
In [3]:
def plotHistDist(func, x, r, title, l, xlabel, ylabel): 
    plt.hist(r, normed=True, histtype='stepfilled', alpha=0.2) 
    plotDist(x, func, title , l, xlabel, ylabel)
#given functions
def plotDist(x, func , title , l , xlabel , ylabel):
    plt.plot(x, func, 'b', lw=2, alpha=0.6, label=l) 
    xl = plt.gca().get_xlim()
    #lines on Y−axis
    plt.hlines(0, xl[0], xl[1], linestyles='--', colors='#999999') 
    plt.gca().set_xlim ( xl )
    plt.legend(loc='best', frameon=False)
    plt.xlabel(xlabel)
    plt.ylabel(ylabel)
    plt.title(title)
In [4]:
#Making dataFrame of counts of attacks in each country 
terrCnt = dataset1.country_txt.value_counts()
dfCnt = pd.DataFrame(terrCnt)
dfCnt = dfCnt.rename(columns={'country_txt':'NumberofAttacks'})

US probability of attacks

In [5]:
#Poisson distribution
tot = 0
arr = [0,0,0,0,0,0,0,0,0]
for i in range(9):
    UnitedStates = df.loc[df['country'] == 217]
    GT1970 = UnitedStates[UnitedStates['iyear'] > 1972+(5*i)]
    fcnt = GT1970[GT1970['iyear'] < 1977+(5*i)]
    fcnt = fcnt.country_txt.value_counts()
    arr[i] = fcnt[0]
    tot += fcnt[0]
    
avg = tot/9
print("The probability there will be exactly 200 attacks in the US within the next five-year period:")
print(stats.poisson.pmf(200, avg))


total = 0
for i in range(200):
    tmp = stats.poisson.pmf(i,avg)
    total += tmp
print("\nThe probability there will be more than 200 attacks in the US within the next five-year period:")
probability = 1 - total
print(probability)

The probability there will be exactly 200 attacks in the US within the next five-year period:
0.008836103705157198

The probability there will be more than 200 attacks in the US within the next five-year period:
0.06685510229484393

Spain probability of attacks

In [6]:
#Poisson distribution
tot = 0
for i in range(9):
    UnitedStates = df.loc[df['country'] == 185]
    GT1970 = UnitedStates[UnitedStates['iyear'] > 1972+(5*i)]
    fcnt = GT1970[GT1970['iyear'] < 1977+(5*i)]
    fcnt = fcnt.country_txt.value_counts()
    tot += fcnt[0]
    arr[i] = fcnt[0]
    
avg = tot/9
arr = np.arange(3000)

print("The probability there will be exactly 200 attacks in Spain within the next five-year period:")
print(stats.poisson.pmf(200, avg))


total = 0
for i in range(250):
    tmp = stats.poisson.pmf(i,avg)
    total += tmp
print("\nThe probability there will be more than 250 attacks in Spain within the next five-year period:")
print(1 - total)

The probability there will be exactly 200 attacks in Spain within the next five-year period:
8.031526999655393e-09

The probability there will be more than 250 attacks in Spain within the next five-year period:
0.9896156867013342

Iran probability of attacks

In [7]:
#Poisson distribution
tot = 0
for i in range(9):
    UnitedStates = df.loc[df['country'] == 94]
    GT1970 = UnitedStates[UnitedStates['iyear'] > 1972+(5*i)]
    fcnt = GT1970[GT1970['iyear'] < 1977+(5*i)]
    fcnt = fcnt.country_txt.value_counts()
    tot += fcnt[0]
    
avg = tot/9
arr = np.arange(3000)

print("The probability there will be exactly 90 attacks in Iran within the next five-year period:")
print(stats.poisson.pmf(90, avg))


total = 0
for i in range(50):
    tmp = stats.poisson.pmf(i,avg)
    total += tmp
print("\nThe probability there will be more than 50 attacks in Iran within the next five-year period:")
print(1 - total)

The probability there will be exactly 90 attacks in Iran within the next five-year period:
0.00046439888630927113

The probability there will be more than 50 attacks in Iran within the next five-year period:
0.9725611492330197

Iraq probability of attacks

In [8]:
#Poisson distribution
tot = 0
arr = [0,0,0,0,0,0,0,0,0]
for i in range(9):
    UnitedStates = df.loc[df['country'] == 95]
    GT1970 = UnitedStates[UnitedStates['iyear'] > 1972+(5*i)]
    fcnt = GT1970[GT1970['iyear'] < 1977+(5*i)]
    #print(GT1970)
    fcnt = fcnt.country_txt.value_counts()
    arr[i] = fcnt[0]
    tot += fcnt[0]
    
avg = tot/9
print("The probability there will be exactly 2100 attacks in Iraq within the next five-year period:")
print(stats.poisson.pmf(2100, avg))


total = 0
for i in range(2000):
    tmp = stats.poisson.pmf(i,avg)
    total += tmp
print("\nThe probability there will be more than 2000 attacks in Iraq within the next five-year period:")
print(1 - total)

The probability there will be exactly 2100 attacks in Iraq within the next five-year period:
0.002001735073177817

The probability there will be more than 2000 attacks in Iraq within the next five-year period:
0.99995349841246
In [9]:
#NorthAmerica Happiness score vs number of attacks
No = df.loc[df['iyear'] > 2013]
NorthAmerica = No.loc[No['region'] == 1]
NorthAmericaTerrCnt = NorthAmerica.country_txt.value_counts()
dfNA = pd.DataFrame(NorthAmericaTerrCnt)
dfNA = dfNA.rename(columns={'country_txt':'NumberofAttacks'})
dfNA
happiness = pd.read_csv('World-Happiness-Report/2017.csv', engine = 'python')
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfNA],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})
combined
slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
graph = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})
plt.show()
print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is -13
Intercept is 182
R value is -0.053140
P value is 0.966154
Std Error is 248

Happiness Score vs Wounded

In [10]:
#Happiness score vs wounded 
df = df.loc[df['iyear'] > 2013]
wound = df.loc[df['region'] == 10]
wound = pd.concat([wound['country_txt'],wound['nwound']],axis=1)
filtered = wound.loc[df['nwound'] > 0]
pwound = filtered.set_index('country_txt')
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
#cant use df because its a different dataset! use Happiness dataset
reducedPwound = pwound.groupby(pwound.index).sum()
wounded = pd.concat([phappiness,reducedPwound],axis=1,join='inner')
print(wounded)
slope, intercept, r_value, p_value, std_err = stats.linregress(wounded['Happiness.Score'],wounded['nwound'])
ax = sns.regplot(x="Happiness.Score", y="nwound", data=wounded, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})
#wounded
print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

              Happiness.Score   nwound
Israel                  7.213    460.0
Qatar                   6.375      1.0
Saudi Arabia            6.344    569.0
Kuwait                  6.105    228.0
Bahrain                 6.087     89.0
Algeria                 5.872     70.0
Libya                   5.525   2651.0
Turkey                  5.500   4292.0
Jordan                  5.336     84.0
Lebanon                 5.225   1002.0
Tunisia                 4.805    216.0
Egypt                   4.735   2854.0
Iran                    4.692     91.0
Iraq                    4.497  49334.0
Yemen                   3.593   6433.0
Syria                   3.462   9277.0
Slope is -4467
Intercept is 28690
R value is -0.372715
P value is 0.155094
Std Error is 2972

NorthAmerica Happiness score vs number of attacks

In [11]:
No = df.loc[df['iyear'] > 2013]
NorthAmerica = No.loc[No['region'] == 1]
NorthAmericaTerrCnt = NorthAmerica.country_txt.value_counts()
dfNA = pd.DataFrame(NorthAmericaTerrCnt)
dfNA = dfNA.rename(columns={'country_txt':'NumberofAttacks'})
dfNA
happiness = pd.read_csv('World-Happiness-Report/2017.csv', engine = 'python')
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfNA],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})
combined

slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
graph = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})
plt.show()

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is -13
Intercept is 182
R value is -0.053140
P value is 0.966154
Std Error is 248

Middle East Happiness score vs number of attacks

In [12]:
MiddleEast = df.loc[df['region'] == 10]
MiddleEastTerrCnt = MiddleEast.country_txt.value_counts()
dfME = pd.DataFrame(MiddleEastTerrCnt)
dfME = dfME.rename(columns={'country_txt':'NumberofAttacks'})
dfME
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfME],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})

slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
ax = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is -1049
Intercept is 6905
R value is -0.361460
P value is 0.140529
Std Error is 676

Sub-Saharan Africa Happiness score vs number of attacks

In [13]:
SS = df.loc[df['region'] == 11]
SSTerrCnt = SS.country_txt.value_counts()
dfSS = pd.DataFrame(SSTerrCnt)
dfSS = dfSS.rename(columns={'country_txt':'NumberofAttacks'})
dfSS
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfSS],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})


slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
ax = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is 487
Intercept is -1735
R value is 0.469961
P value is 0.007638
Std Error is 169

Western Europe Happiness score vs number of attacks

In [14]:
WE = df.loc[df['region'] == 8]
WETerrCnt = WE.country_txt.value_counts()
dfWE = pd.DataFrame(WETerrCnt)
dfWE = dfWE.rename(columns={'country_txt':'NumberofAttacks'})
dfWE
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfWE],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})

slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
ax = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is -32
Intercept is 285
R value is -0.210182
P value is 0.402526
Std Error is 38

Southeast Asia Happiness score vs number of attacks

In [15]:
SEA = df.loc[df['region'] == 5]
SEATerrCnt = SEA.country_txt.value_counts()
dfSEA = pd.DataFrame(SEATerrCnt)
dfSEA = dfSEA.rename(columns={'country_txt':'NumberofAttacks'})
dfSEA
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfSEA],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})


slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
ax = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is 352
Intercept is -1220
R value is 0.229888
P value is 0.661242
Std Error is 746

South America Happiness score vs number of attacks

In [16]:
SA = df.loc[df['region'] == 3]
SATerrCnt = SA.country_txt.value_counts()
dfSA = pd.DataFrame(SATerrCnt)
dfSA = dfSA.rename(columns={'country_txt':'NumberofAttacks'})
dfSA
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfSA],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})

slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
ax = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is 40
Intercept is -159
R value is 0.112695
P value is 0.772835
Std Error is 134

Eastern Europe Happiness score vs number of attacks

In [17]:
EE = df.loc[df['region'] == 9]
EETerrCnt = EE.country_txt.value_counts()
dfEE = pd.DataFrame(EETerrCnt)
dfEE = dfEE.rename(columns={'country_txt':'NumberofAttacks'})
dfEE
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')
combined = pd.concat([phappiness,dfEE],axis=1,join='inner')
combined = combined.rename(columns={'country_txt':'NumberofAttacks'})

slope, intercept, r_value, p_value, std_err = stats.linregress(combined['Happiness.Score'],combined['NumberofAttacks'])
ax = sns.regplot(x="Happiness.Score", y="NumberofAttacks", data=combined, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

Slope is -385
Intercept is 2220
R value is -0.565177
P value is 0.035195
Std Error is 162

Happiness score vs wounded

In [18]:
df = df.loc[df['iyear'] > 2013]
wound = df.loc[df['region'] == 10]
wound = pd.concat([wound['country_txt'],wound['nwound']],axis=1)
filtered = wound.loc[df['nwound'] > 0]
pwound = filtered.set_index('country_txt')
Happiness = pd.concat([happiness['Country'],happiness['Happiness.Score']],axis=1)
phappiness = Happiness.set_index('Country')

#cant use df because its a different dataset! use Happiness dataset
reducedPwound = pwound.groupby(pwound.index).sum()
wounded = pd.concat([phappiness,reducedPwound],axis=1,join='inner')
print(wounded)
slope, intercept, r_value, p_value, std_err = stats.linregress(wounded['Happiness.Score'],wounded['nwound'])
ax = sns.regplot(x="Happiness.Score", y="nwound", data=wounded, color='b', line_kws={'label':"y={0:.1f}x+{1:.1f}".format(slope,intercept)})

#wounded
print("Slope is %d" % slope)
print("Intercept is %d" % intercept)
print("R value is %f" % r_value)
print("P value is %f" % p_value)
print("Std Error is %d" % std_err)

              Happiness.Score   nwound
Israel                  7.213    460.0
Qatar                   6.375      1.0
Saudi Arabia            6.344    569.0
Kuwait                  6.105    228.0
Bahrain                 6.087     89.0
Algeria                 5.872     70.0
Libya                   5.525   2651.0
Turkey                  5.500   4292.0
Jordan                  5.336     84.0
Lebanon                 5.225   1002.0
Tunisia                 4.805    216.0
Egypt                   4.735   2854.0
Iran                    4.692     91.0
Iraq                    4.497  49334.0
Yemen                   3.593   6433.0
Syria                   3.462   9277.0
Slope is -4467
Intercept is 28690
R value is -0.372715
P value is 0.155094
Std Error is 2972

Happiness Of Nations

In [19]:
import pandas as pd
import numpy as np
from pandas import Series, DataFrame
import matplotlib.pyplot as plt
import matplotlib

1. Import Dataset

In [20]:
gbterr = df.copy()
#focus on terrorist attacks which happened after 2010 first because that is around the time the world happiness report gets its data
gbterr2017 = gbterr[gbterr.iyear >2013]
happiness = pd.read_csv('World-Happiness-Report/2017.csv', engine = 'python', index_col=0)

2. Visualization

In [21]:
#lets visualize happiness scores 
mng = plt.get_current_fig_manager()
mng.show_popup("True")

/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:3: MatplotlibDeprecationWarning: The show_popup function was deprecated in version 2.2.
  This is separate from the ipykernel package so we can avoid doing imports until

<Figure size 432x288 with 0 Axes>
In [22]:
fig = plt.figure()
fig.set_size_inches(50,50)
happiness['Happiness.Score'].plot('bar')
plt.show()
In [23]:
matplotlib.rcParams.update({'font.size': 35})
fig = plt.figure()
fig.set_size_inches(50,100)
happiness['Happiness.Score'].plot('barh')
Out[23]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0af55e80>
In [24]:
#now from the terrorism 2017 dataframe we get the number of attacks per country
tattacks_count = gbterr2017['country_txt'].value_counts()
tattacks_count.head()
Out[24]:
Iraq           12510
Afghanistan     6783
Pakistan        4977
India           3735
Philippines     2642
Name: country_txt, dtype: int64
In [25]:
#bar graph of terrorism attacks per country in 2017
fig = plt.figure()
matplotlib.rcParams.update({'font.size': 15})
fig.set_size_inches(40,30)
tattacks_count.plot("bar")
plt.show()

3. Correlations

In [26]:
#now we look for a correlation between the number terrorist attacks in 2017 in a country and that country's happiness score as well as other variables in the happiness database
#tattacks_count
h_score = pd.Series(happiness['Happiness.Score'])
#h_score.index.name= "Country"
tattacks_count.index.name = "Country"
tattacks_count = pd.DataFrame(tattacks_count)
tattacks_count = tattacks_count.rename(columns={"country_txt":"NumOfAttacks"})
In [27]:
a_h =  pd.merge(happiness,tattacks_count, on='Country')
dfcorr = a_h[["Happiness.Score", "NumOfAttacks","Economy..GDP.per.Capita.", "Freedom","Trust..Government.Corruption.","Dystopia.Residual","Family","Health..Life.Expectancy." ]]
dfcorr.corr()
Out[27]:
Happiness.Score NumOfAttacks Economy..GDP.per.Capita. Freedom Trust..Government.Corruption. Dystopia.Residual Family Health..Life.Expectancy.
Happiness.Score 1.000000 -0.169753 0.834426 0.619353 0.461512 0.499159 0.769330 0.786620
NumOfAttacks -0.169753 1.000000 -0.111607 -0.169398 -0.079760 -0.004880 -0.250197 -0.166409
Economy..GDP.per.Capita. 0.834426 -0.111607 1.000000 0.383448 0.348569 0.109302 0.702235 0.834958
Freedom 0.619353 -0.169398 0.383448 1.000000 0.553646 0.145555 0.499936 0.371297
Trust..Government.Corruption. 0.461512 -0.079760 0.348569 0.553646 1.000000 0.055068 0.252029 0.272360
Dystopia.Residual 0.499159 -0.004880 0.109302 0.145555 0.055068 1.000000 0.101564 0.108849
Family 0.769330 -0.250197 0.702235 0.499936 0.252029 0.101564 1.000000 0.616311
Health..Life.Expectancy. 0.786620 -0.166409 0.834958 0.371297 0.272360 0.108849 0.616311 1.000000
In [28]:
f,ax = plt.subplots(figsize=(25, 18))
sns.heatmap(dfcorr.corr(), annot=True, linewidths=.5, fmt= '.1f',ax=ax)
plt.show()

Our initial hypothesis was that the number of terrorist attacks a country suffered would have some correlation with that countries hapiness score.

From the correlations shown above, we didn't find substantially large correlations between number of attacks and these variables.

However we did find small negative correlations between the number of attacks and most of the variables that influence happiness in countries.(Except for Dystopia score which had no correlation whatsover). However, the correlations are not as big as we expected in our initial hypothesis.

We decided to look at other data sets in order to find other factors that influence a country's happiness score

UN Country Profiles 2017

This dataset contains key statistical indicators of the countries. It covers 4 major sections

-General Information, Economic Indicators, Social Indicators, Environmental & Infrastructure Indicators

In [29]:
un_report =  pd.read_csv('undata-country-profiles/country_profile_variables.csv', engine = 'python', index_col= 0)
un_report.index.name = "Country"
un_report.head()
Out[29]:
Region Surface area (km2) Population in thousands (2017) Population density (per km2, 2017) Sex ratio (m per 100 f, 2017) GDP: Gross domestic product (million current US$) GDP growth rate (annual %, const. 2005 prices) GDP per capita (current US$) Economy: Agriculture (% of GVA) Economy: Industry (% of GVA) ... Mobile-cellular subscriptions (per 100 inhabitants).1 Individuals using the Internet (per 100 inhabitants) Threatened species (number) Forested area (% of land area) CO2 emission estimates (million tons/tons per capita) Energy production, primary (Petajoules) Energy supply per capita (Gigajoules) Pop. using improved drinking water (urban/rural, %) Pop. using improved sanitation facilities (urban/rural, %) Net Official Development Assist. received (% of GNI)
Country
Afghanistan SouthernAsia 652864 35530 54.4 106.3 20270 -2.4 623.2 23.3 23.3 ... 8.3 42 2.1 9.8/0.3 63 5 78.2/47.0 45.1/27.0 21.43 -99
Albania SouthernEurope 28748 2930 106.9 101.9 11541 2.6 3984.2 22.4 26.0 ... 63.3 130 28.2 5.7/2.0 84 36 94.9/95.2 95.5/90.2 2.96 -99
Algeria NorthernAfrica 2381741 41318 17.3 102.0 164779 3.8 4154.1 12.2 37.3 ... 38.2 135 0.8 145.4/3.7 5900 55 84.3/81.8 89.8/82.2 0.05 -99
American Samoa Polynesia 199 56 278.2 103.6 -99 -99 -99.0 -99 -99.0 ... -99 92 87.9 -99 -99 -99 100.0/100.0 62.5/62.5 -99 -99
Andorra SouthernEurope 468 77 163.8 102.3 2812 0.8 39896.4 0.5 10.8 ... 96.9 13 34.0 0.5/6.4 1 119 100.0/100.0 100.0/100.0 -99 -99

5 rows × 49 columns

In order to better visualize our data we decided to group countries into groups according to their GDP

We will be using the World Bank's classfication of low income, lower middle income, upper middle income, and high-income.

#Countries with less than $1,035 GNI per capita are classified as low-income countries, those with between $1,036 and $4,085 as lower middle income countries, those with between $4,086 and $12,615 as upper middle income countries, and those with incomes of more than $12,615 as high-income countries.
In [30]:
un_report["GDP per capita (current US$)"].head()
Out[30]:
Country
Afghanistan         623.2
Albania            3984.2
Algeria            4154.1
American Samoa      -99.0
Andorra           39896.4
Name: GDP per capita (current US$), dtype: float64

Low-Income Countries ( GDP per capita < 1035)

In [31]:
low_income = un_report[un_report["GDP per capita (current US$)"] < 1035]
li_gdpc = low_income[low_income["GDP per capita (current US$)"] >0]
li_gdpc  = li_gdpc["GDP per capita (current US$)"]
li_gdpc = li_gdpc.sort_values()

Bar Graph: Low Income countries and their GDP

In [32]:
matplotlib.rcParams.update({'font.size': 30})
fig = plt.figure()
fig.set_size_inches(50,20)
li_gdpc.plot("bar")
Out[32]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0af61f60>

Happiness Scores of Low Income Countries

In [33]:
li_gdpc.index
h_score.index
h_score_lowincome = h_score[li_gdpc.index]
h_score_lowincome = h_score_lowincome.dropna()
h_score_lowincome = h_score_lowincome.sort_values()
h_score_lowincome.head()
Out[33]:
Country
Central African Republic    2.693
Burundi                     2.905
Rwanda                      3.471
Togo                        3.495
Guinea                      3.507
Name: Happiness.Score, dtype: float64
In [34]:
h_score_lowincome
Out[34]:
Country
Central African Republic    2.693
Burundi                     2.905
Rwanda                      3.471
Togo                        3.495
Guinea                      3.507
Liberia                     3.533
Haiti                       3.603
Madagascar                  3.644
Benin                       3.657
Afghanistan                 3.794
Lesotho                     3.808
Zimbabwe                    3.875
Chad                        3.936
Malawi                      3.970
Niger                       4.028
Burkina Faso                4.032
Uganda                      4.081
Mali                        4.190
Ethiopia                    4.460
Senegal                     4.535
Mozambique                  4.550
Sierra Leone                4.709
Nepal                       4.962
Tajikistan                  5.041
Somalia                     5.151
Name: Happiness.Score, dtype: float64
In [35]:
matplotlib.rcParams.update({'font.size': 30})
fig = plt.figure()
fig.set_size_inches(50,20)
h_score_lowincome.plot("bar")
Out[35]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b09b3c8>
In [36]:
h_score_lowincome.describe()
Out[36]:
count    25.000000
mean      3.985200
std       0.615126
min       2.693000
25%       3.603000
50%       3.936000
75%       4.460000
max       5.151000
Name: Happiness.Score, dtype: float64

Terrorist Attacks in Low Income Countries

In [37]:
tattacks_count = gbterr2017['country_txt'].value_counts()
tattacks_count.index.name = "Country"
tattacks_count.name = "Terrorist Attacks Count"
In [38]:
tattacks_lowincome = tattacks_count[h_score_lowincome.index]
tattacks_lowincome = tattacks_lowincome.dropna()
tattacks_lowincome = tattacks_lowincome.sort_values()
tattacks_lowincome.head()
Out[38]:
Country
Malawi        1.0
Haiti         1.0
Lesotho       1.0
Liberia       2.0
Madagascar    2.0
Name: Terrorist Attacks Count, dtype: float64
In [39]:
matplotlib.rcParams.update({'font.size': 30})
fig = plt.figure()
fig.set_size_inches(50,20)
tattacks_lowincome.plot("bar")
Out[39]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b5e6cf8>
In [40]:
tattacks_lowincome.describe()
Out[40]:
count      23.000000
mean      474.652174
std      1469.930382
min         1.000000
25%         3.500000
50%        39.000000
75%       165.500000
max      6783.000000
Name: Terrorist Attacks Count, dtype: float64

Relationship between the number Terrorism Attacks and Happiness Score of Low Income Countries

In [41]:
tattacks_lowincome.corr(h_score_lowincome)
Out[41]:
0.06151287918229158

Relationship between the number Terrorism Attacks and Happiness Score other hapiness variables of Low Income Countries

In [42]:
lowincomea_h = happiness.copy()
lowincomea_h["Attacks"] = tattacks_lowincome
lowincomea_h.corr()
Out[42]:
Happiness.Rank Happiness.Score Whisker.high Whisker.low Economy..GDP.per.Capita. Family Health..Life.Expectancy. Freedom Generosity Trust..Government.Corruption. Dystopia.Residual Attacks
Happiness.Rank 1.000000 -0.992774 -0.993058 -0.991533 -0.813244 -0.736753 -0.780716 -0.551608 -0.132620 -0.405842 -0.484506 -0.047867
Happiness.Score -0.992774 1.000000 0.999497 0.999520 0.812469 0.752737 0.781951 0.570137 0.155256 0.429080 0.475355 0.061513
Whisker.high -0.993058 0.999497 1.000000 0.998036 0.811868 0.750934 0.776634 0.569907 0.155462 0.426459 0.478824 0.049375
Whisker.low -0.991533 0.999520 0.998036 1.000000 0.812267 0.753767 0.786385 0.569808 0.154904 0.431223 0.471505 0.072913
Economy..GDP.per.Capita. -0.813244 0.812469 0.811868 0.812267 1.000000 0.688296 0.843077 0.369873 -0.019011 0.350944 0.024226 -0.033365
Family -0.736753 0.752737 0.750934 0.753767 0.688296 1.000000 0.612080 0.424966 0.051693 0.231841 0.070506 -0.268013
Health..Life.Expectancy. -0.780716 0.781951 0.776634 0.786385 0.843077 0.612080 1.000000 0.349827 0.063191 0.279752 0.054963 -0.123284
Freedom -0.551608 0.570137 0.569907 0.569808 0.369873 0.424966 0.349827 1.000000 0.316083 0.499183 0.081926 -0.196741
Generosity -0.132620 0.155256 0.155462 0.154904 -0.019011 0.051693 0.063191 0.316083 1.000000 0.294159 -0.116627 0.148426
Trust..Government.Corruption. -0.405842 0.429080 0.426459 0.431223 0.350944 0.231841 0.279752 0.499183 0.294159 1.000000 -0.022755 -0.011623
Dystopia.Residual -0.484506 0.475355 0.478824 0.471505 0.024226 0.070506 0.054963 0.081926 -0.116627 -0.022755 1.000000 0.331485
Attacks -0.047867 0.061513 0.049375 0.072913 -0.033365 -0.268013 -0.123284 -0.196741 0.148426 -0.011623 0.331485 1.000000
In [43]:
f,ax = plt.subplots(figsize=(25, 18))
sns.heatmap(lowincomea_h.corr(), annot=True, linewidths=.5, fmt= '.1f',ax=ax)
plt.show()

Lower Middle Income Countries (GDP per Capita between 1,036 and 4,085)

In [44]:
low_middle_income = un_report[un_report["GDP per capita (current US$)"] > 1035  ]
low_middle_income = low_middle_income[low_middle_income["GDP per capita (current US$)"] <4085]
lmi_gdpc = low_middle_income["GDP per capita (current US$)"]
lmi_gdpc = lmi_gdpc.sort_values()
matplotlib.rcParams.update({'font.size': 40})
In [45]:
fig = plt.figure()
fig.set_size_inches(50,20)
lmi_gdpc.plot("bar")
Out[45]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b6619b0>
In [46]:
lmi_gdpc.describe()
Out[46]:
count      53.000000
mean     2390.367925
std       961.348547
min      1067.000000
25%      1442.900000
50%      2308.300000
75%      3211.700000
max      4006.000000
Name: GDP per capita (current US$), dtype: float64

Happiness Scores of Lower Middle Income Countries

In [47]:
matplotlib.rcParams.update({'font.size': 10})
lmi_hscore = h_score[lmi_gdpc.index]
lmi_hscore = lmi_hscore.sort_values()
lmi_hscore =lmi_hscore.dropna()
lmi_hscore.head()
lmi_hscore.name= "Happiness Score"
In [48]:
matplotlib.rcParams.update({'font.size': 30})
fig = plt.figure()
fig.set_size_inches(50,20)
lmi_hscore.plot("bar")
Out[48]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b7e3048>
In [49]:
lmi_hscore.describe()
Out[49]:
count    31.000000
mean      4.755871
std       0.660202
min       3.591000
25%       4.303500
50%       4.644000
75%       5.127500
max       6.454000
Name: Happiness Score, dtype: float64

Terrorist Attacks in Lower Middle Income Countries

In [50]:
tattacks_lmi = tattacks_count[lmi_hscore.index]
tattacks_lmi = tattacks_lmi.dropna()
tattacks_lmi =tattacks_lmi.sort_values()
In [51]:
matplotlib.rcParams.update({'font.size': 30})
fig = plt.figure()
fig.set_size_inches(50,20)
tattacks_lmi.plot("bar")
Out[51]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0af9c4a8>
In [52]:
tattacks_lmi.describe()
Out[52]:
count      27.000000
mean      809.407407
std      1313.870508
min         1.000000
25%         3.500000
50%       110.000000
75%      1164.500000
max      4977.000000
Name: Terrorist Attacks Count, dtype: float64

Relationship between the number Terrorism Attacks and Happiness Score of Lower Middle Income Countries

In [53]:
tattacks_lmi.head()
Out[53]:
Country
Morocco       1.0
Bhutan        1.0
Uzbekistan    1.0
Nicaragua     2.0
Ghana         2.0
Name: Terrorist Attacks Count, dtype: float64
In [54]:
lmi_hscore.head()
Out[54]:
Country
South Sudan    3.591
Yemen          3.593
Ukraine        4.096
Ghana          4.120
Sudan          4.139
Name: Happiness Score, dtype: float64
In [55]:
lmi_hscore.corr(tattacks_lmi)
Out[55]:
-0.016847750282220065
In [56]:
lowerMiddleincomea_h = happiness.copy()
lowerMiddleincomea_h["Attacks"] = tattacks_lmi
In [57]:
f,ax = plt.subplots(figsize=(25, 18))
sns.heatmap(lowerMiddleincomea_h.corr(), annot=True, linewidths=.5, fmt= '.1f',ax=ax)
plt.show()

Upper Middle Income Countries (GDP per capita 4,086 and 12,615 )

In [58]:
upper_middle_income = un_report[un_report["GDP per capita (current US$)"] > 4086  ]
upper_middle_income = upper_middle_income[upper_middle_income["GDP per capita (current US$)"] <12615]
umi_gdpc = upper_middle_income["GDP per capita (current US$)"]
umi_gdpc = umi_gdpc.sort_values()
matplotlib.rcParams.update({'font.size': 40})
In [59]:
fig = plt.figure()
fig.set_size_inches(50,20)
umi_gdpc.plot("bar")
Out[59]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0aebdbe0>
In [60]:
umi_gdpc.describe()
Out[60]:
count       53.000000
mean      7298.392453
std       2448.946058
min       4154.100000
25%       5105.800000
50%       6739.200000
75%       9040.900000
max      12355.500000
Name: GDP per capita (current US$), dtype: float64
In [61]:
matplotlib.rcParams.update({'font.size': 20})

Happiness Scores of Upper Middle Income Countries

In [62]:
umi_hscore = h_score[umi_gdpc.index]
umi_hscore = umi_hscore.dropna()
umi_hscore = umi_hscore.sort_values()
In [63]:
fig = plt.figure()
fig.set_size_inches(30,20)
umi_hscore.plot("bar")
Out[63]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0adc9908>
In [64]:
umi_hscore.describe()
Out[64]:
count    38.000000
mean      5.488053
std       0.713442
min       3.766000
25%       5.231000
50%       5.496500
75%       5.935000
max       7.079000
Name: Happiness.Score, dtype: float64

Terrorist Attacks in Lower Middle Income Countries

In [65]:
tattacks_umi = tattacks_count[umi_hscore.index]
tattacks_umi = tattacks_umi.dropna()
tattacks_umi = tattacks_umi.sort_values()
In [66]:
matplotlib.rcParams.update({'font.size': 15})
fig = plt.figure()
fig.set_size_inches(20,20)
tattacks_umi.plot("bar")
Out[66]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b8b3550>
In [67]:
tattacks_umi.describe()
Out[67]:
count       28.000000
mean       650.250000
std       2370.149919
min          1.000000
25%          3.000000
50%         18.000000
75%         69.000000
max      12510.000000
Name: Terrorist Attacks Count, dtype: float64

Relationship between the number Terrorism Attacks and Happiness Score of Upper Middle Income Countries

In [68]:
tattacks_umi.head()
Out[68]:
Country
Serbia                1.0
Turkmenistan          1.0
Montenegro            1.0
Dominican Republic    2.0
Poland                3.0
Name: Terrorist Attacks Count, dtype: float64
In [69]:
umi_hscore.head()
Out[69]:
Country
Botswana    3.766
Angola      3.795
Gabon       4.465
Iraq        4.497
Namibia     4.574
Name: Happiness.Score, dtype: float64
In [70]:
tattacks_umi.corr(umi_hscore)
Out[70]:
-0.25404708132522136
In [71]:
upperMiddleincomea_h = happiness.copy()
upperMiddleincomea_h["Attacks"] = tattacks_umi
upperMiddleincomea_h = upperMiddleincomea_h.dropna()
upperMiddleincomea_h
Out[71]:
Happiness.Rank Happiness.Score Whisker.high Whisker.low Economy..GDP.per.Capita. Family Health..Life.Expectancy. Freedom Generosity Trust..Government.Corruption. Dystopia.Residual Attacks
Country
Brazil 22 6.635 6.725470 6.544531 1.107353 1.431306 0.616552 0.437454 0.162350 0.111093 2.769267 12.0
Mexico 25 6.578 6.671149 6.484851 1.153184 1.210862 0.709979 0.412730 0.120990 0.132774 2.837155 49.0
Thailand 32 6.424 6.509117 6.338883 1.127869 1.425792 0.647239 0.580201 0.572123 0.031613 2.039508 1209.0
Colombia 36 6.357 6.452020 6.261980 1.070622 1.402183 0.595028 0.477487 0.149014 0.046669 2.616068 593.0
Malaysia 42 6.084 6.179980 5.988021 1.291215 1.284646 0.618784 0.402265 0.416609 0.065601 2.004449 40.0
Ecuador 44 6.008 6.105848 5.910152 1.000820 1.286169 0.685636 0.455198 0.150112 0.140135 2.290353 6.0
Poland 46 5.973 6.053908 5.892092 1.291788 1.445712 0.699475 0.520342 0.158466 0.059308 1.797723 3.0
Algeria 53 5.872 5.978286 5.765714 1.091864 1.146217 0.617585 0.233336 0.069437 0.146096 2.567604 52.0
Turkmenistan 59 5.822 5.885181 5.758819 1.130777 1.493149 0.437726 0.418272 0.249925 0.259270 1.832910 1.0
Kazakhstan 60 5.819 5.903642 5.734358 1.284556 1.384369 0.606042 0.437454 0.201964 0.119283 1.784893 3.0
Peru 63 5.715 5.811947 5.618054 1.035225 1.218770 0.630166 0.450003 0.126820 0.047049 2.207269 33.0
Libya 68 5.525 5.676954 5.373046 1.101803 1.357564 0.520169 0.465733 0.152074 0.092610 1.835011 1883.0
Turkey 69 5.500 5.594865 5.405135 1.198274 1.337753 0.637606 0.300741 0.046693 0.099672 1.879278 1239.0
Paraguay 70 5.493 5.577381 5.408619 0.932537 1.507285 0.579251 0.473508 0.224151 0.091066 1.685333 59.0
Serbia 73 5.395 5.491570 5.298430 1.069318 1.258190 0.650785 0.208716 0.220126 0.040904 1.947084 1.0
Jordan 74 5.336 5.448410 5.223590 0.991012 1.239089 0.604590 0.418421 0.172170 0.119803 1.791177 24.0
Hungary 75 5.324 5.403040 5.244960 1.286012 1.343133 0.687763 0.175864 0.078402 0.036637 1.716459 3.0
Jamaica 76 5.311 5.581399 5.040601 0.925579 1.368218 0.641022 0.474307 0.233818 0.055268 1.612326 3.0
China 79 5.273 5.319278 5.226721 1.081166 1.160837 0.741416 0.472788 0.028807 0.022794 1.764939 65.0
Montenegro 83 5.237 5.341044 5.132956 1.121129 1.238376 0.667465 0.194989 0.197911 0.088174 1.729192 1.0
Azerbaijan 85 5.234 5.299287 5.168714 1.153602 1.152400 0.540776 0.398156 0.045269 0.180988 1.762482 7.0
Dominican Republic 86 5.230 5.349061 5.110939 1.079374 1.402417 0.574874 0.552590 0.186968 0.113945 1.319465 2.0
Lebanon 88 5.225 5.318882 5.131118 1.074988 1.129624 0.735081 0.288516 0.264451 0.037514 1.695074 312.0
South Africa 101 4.829 4.929435 4.728565 1.054699 1.384789 0.187080 0.479247 0.139362 0.072509 1.510909 81.0
Bulgaria 105 4.714 4.803695 4.624306 1.161459 1.434379 0.708218 0.289232 0.113178 0.011052 0.996139 4.0
Iraq 117 4.497 4.622591 4.371409 1.102710 0.978613 0.501180 0.288556 0.199637 0.107216 1.318907 12510.0
Gabon 118 4.465 4.557362 4.372639 1.198210 1.155620 0.356579 0.312329 0.043785 0.076047 1.322916 4.0
Angola 140 3.795 3.951642 3.638358 0.858428 1.104412 0.049869 0.000000 0.097926 0.069720 1.614482 8.0
In [72]:
f,ax = plt.subplots(figsize=(25, 18))
sns.heatmap(upperMiddleincomea_h.corr(), annot=True, linewidths=.5, fmt= '.1f',ax=ax)
plt.show()
#note that the only important line in the heat map above is the last line.

Found Negative Correlation

Found Small Negative Correlation in Lower Middle Income Countries

High-Income Countries ( GDP per capita > 12,615)

In [73]:
high_income = un_report[un_report["GDP per capita (current US$)"] > 12615]
hi_gdpc = high_income["GDP per capita (current US$)"]
In [74]:
hi_gdpc = hi_gdpc.sort_values()
In [75]:
matplotlib.rcParams.update({'font.size': 40})
fig = plt.figure()
fig.set_size_inches(50,50)
hi_gdpc.plot("bar")
Out[75]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b8b6b70>
In [76]:
hi_gdpc.describe()
Out[76]:
count        71.000000
mean      38491.988732
std       29897.125398
min       13268.100000
25%       18315.100000
50%       30144.500000
75%       44247.250000
max      169491.800000
Name: GDP per capita (current US$), dtype: float64

Happiness Scores of High-Income Income Countries

In [77]:
highIncome_hscore = h_score[hi_gdpc.index]
highIncome_hscore["United States"] = h_score["United States"]
highIncome_hscore["Czech Republic"] = h_score["Czech Republic"]
highIncome_hscore["South Korea"] = h_score["South Korea"]
highIncome_hscore = highIncome_hscore.dropna()
highIncome_hscore = highIncome_hscore.sort_values()
In [78]:
matplotlib.rcParams.update({'font.size': 30})
fig = plt.figure()
fig.set_size_inches(50,50)
highIncome_hscore.plot("bar")
Out[78]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0af0df60>
In [79]:
highIncome_hscore.describe()
Out[79]:
count    44.000000
mean      6.571136
std       0.645832
min       5.195000
25%       6.095250
50%       6.585500
75%       7.057750
max       7.537000
Name: Happiness.Score, dtype: float64

Terrorist Attacks in High-Income Income Countries

In [80]:
tattacks_hi = tattacks_count[highIncome_hscore.index]
tattacks_hi = tattacks_hi.dropna()
tattacks_hi = tattacks_hi.sort_values()
In [81]:
matplotlib.rcParams.update({'font.size': 20})
fig = plt.figure()
fig.set_size_inches(30,20)
tattacks_hi.plot("bar")
Out[81]:
<matplotlib.axes._subplots.AxesSubplot at 0x1c0b6947f0>
In [82]:
tattacks_hi.describe()
Out[82]:
count     38.000000
mean      60.052632
std      110.710957
min        1.000000
25%        2.000000
50%       11.000000
75%       65.250000
max      444.000000
Name: Terrorist Attacks Count, dtype: float64

Relationship between the number Terrorism Attacks and Happiness Score of High-Income Countries

In [83]:
tattacks_hi.head()
Out[83]:
Country
Norway         1.0
Uruguay        1.0
Latvia         1.0
Iceland        1.0
Switzerland    1.0
Name: Terrorist Attacks Count, dtype: float64
In [84]:
highIncome_hscore.head()
Out[84]:
Country
Portugal    5.195
Greece      5.227
Estonia     5.611
Cyprus      5.621
Slovenia    5.758
Name: Happiness.Score, dtype: float64
In [85]:
tattacks_hi.corr(highIncome_hscore)
Out[85]:
0.04895969625935482
In [86]:
Highincomea_h = happiness.copy()
Highincomea_h["Attacks"] = tattacks_hi
Highincomea_h = Highincomea_h.dropna()
In [ ]:
f,ax = plt.subplots(figsize=(25, 18))
sns.heatmap(Highincomea_h.corr(), annot=True, linewidths=.5, fmt= '.1f',ax=ax)
plt.show()
In [ ]:
tattacks_hi.sum()
tattacks_umi.sum()
tattacks_lmi.sum()
tattacks_lowincome.sum()
In [ ]:
classes = pd.Series(index=["Low Income","Lower Middle Income","Upper Middle Income", "High Income" ])
classes.index.name="Classification"
classes.name="Total Terrorist Attacks"
classes["Low Income"] = tattacks_lowincome.sum()
classes["Lower Middle Income"] = tattacks_lmi.sum() 
classes["Upper Middle Income"] = tattacks_umi.sum() - tattacks_count["Iraq"] #leave or remove Iraq???
classes["High Income"] = tattacks_hi.sum()
classes
In [ ]:
classes.plot("bar")

High Income Countries Stats

Happiness from 5.195(Portugal) to 7.537(Norway)

Mean Happiness = 6.57

Standard Deviation = 0.6458316134876863

In [ ]:
highIncome_hscore.mean()
In [ ]:
highIncome_hscore.std()
In [ ]:
from bokeh.io import output_file, output_notebook, show
from bokeh.io import output_file, show
from bokeh.models import ColumnDataSource, GMapOptions
from bokeh.plotting import gmap

Visualtization: Google Maps API

Shows terrorism attacks across the world.

In [ ]:
df = pd.read_csv('globalTerrorism.csv', engine = 'python')
In [ ]:
latitude_list = df['latitude'].tolist()
longitude_list = df['longitude'].tolist()
In [ ]:
map_options = GMapOptions(lat=29, lng=-98, map_type="roadmap", zoom=7)

p = gmap("AIzaSyAx7GNIhrPKKX7LPlC84iGc3Mg-8_uKdbI", map_options, title="Terrorist Attacks")

source = ColumnDataSource(
    data=dict(lat=df['latitude'].tolist(),
              lon=df['longitude'].tolist())
)

p.circle(x="lon", y="lat", size=15, fill_color="blue", fill_alpha=0.8, source=source)
output_notebook()
show(p)